Surface oxidation and thermoelectric properties of indium-doped tin
telluride nanowires
Z. Li,a E. Z. Xu,a Y. Losovyj,b N. Li,c B. Swartzentruber,d N. Sinitsyn,e J. K. Yoo,c Q. X. Jiac,f and
S. X. Zhang*a
a. Department of Physics,
[email protected].
Indiana
University,
Bloomington,
IN
47405,
USA,
E-mail:
b. Department of Chemistry, Indiana University, Bloomington, IN 47405, USA
c. Center for Integrated Nanotechnologies, Los Alamos National Laboratory, Los Alamos, NM 87545,
USA
d. Center for Integrated Nanotechnologies, Sandia National Laboratories, Albuquerque, NM 87185,
USA
e. Theoretical Division, Los Alamos National Laboratory, Los Alamos, NM 87545, USA
f. Department of Materials Design and Innovation, University at Buffalo – The State University of New
York, Buffalo, NY 14260, USA
Abstract
The recent discovery of topological surface states and the realization of excellent
thermoelectric properties in the SnTe-based compounds have triggered growing interest in the
areas of condensed matter physics and materials science. Indium doping in SnTe is of particular
interest because, depending on the doping level, it can either induce superconductivity that
coexists with topological surface states, or generate resonant states in the bulk valence band
leading to enhanced thermoelectric properties. Here we report the growth of In-doped SnTe
nanowires via vapor deposition and the study of their surface oxidation and thermoelectric
properties. The nanowire growth is assisted by Au catalysts, and their morphologies vary as a
function of substrate position and temperature. Transmission electron microscopy
characterization reveals the formation of an amorphous layer on the surface of the single
crystalline nanowires. X-ray photoelectron spectroscopy studies suggest that the nanowire
surface is composed of In2O3, SnO2, Te and TeO2 which can be readily removed by argon ion
sputtering. Exposure of the cleaned nanowires to atmosphere yields rapid oxidation of the
surface within only one minute. Compared to undoped SnTe, the indium-doped nanowire
shows suppressed electrical and thermal conductivities but enhanced thermopower, yielding an
improved thermoelectric figure of merit ZT.
Introduction
Tin telluride (SnTe)-based narrow band-gap semiconductors have attracted growing
interest owning to the recent discovery of topological surface states1-7 and the realization
of excellent thermoelectric properties.8-10 The topological states in SnTe are
characterized by even number of gapless Dirac cones that reside in the Brillouin zone
of the surfaces satisfying crystal mirror symmetry.1,
3, 4, 11, 12
Breaking the crystal
symmetry via external knobs such as strain or temperature can effectively tune the
topological surface properties, offering potential applications in tunable electronics and
spintronics.13-15 When doped with a few percent of indium, SnTe becomes a
superconductor below its critical temperature.16-20 Angle-resolved photoemission
spectroscopy (ARPES) studies have revealed the signature of topological surface
states.21 The coexistence of surface states and superconductivity may facilitate the
realization of the long-sought particles named Majorana fermions which serve as the
basis for topological quantum computing.22
Beyond topological states, SnTe-based compounds have also been demonstrated to
be promising thermoelectric (TE) materials that are more environmentally friendly than
the conventional lead chalcogenides.8-10, 23-29 The TE figure of merit is defined by a
dimensionless parameter: 𝑍𝑇 =
𝑆 2 𝜎𝑇
𝜅
, where S is the thermopower or Seebeck
coefficient, σ the electrical conductivity and κ the thermal conductivity. While the
pristine bulk SnTe crystal is not compelling because of its low thermopower and high
2
thermal conductivity,30 doped and alloyed SnTe exhibits significantly enhanced
thermoelectric performance.8,
9, 31-36
In particular, less than 2% indium-doping can
induce resonant states in the valence band of SnTe, which dramatically improve its
thermopower and hence the ZT.8 The peak ZT achieved in In-doped SnTe is as high as
1.1 at 873 K, suggesting its great potential to replace lead chalcogenides for applications
in waste heater recovery.8
In comparison to bulk crystals, nanostructures such as nanoplates and nanowires
provide a unique platform to better exploit topological properties and to further improve
thermoelectric performance. From a topological state point of view, the large surfacearea-to-volume ratio of nanostructures magnifies contributions from the surface states
over the bulk states of the material.37-42 The topological surface could have enhanced
thermopower due to its unique energy dispersion and strongly energy-dependent
relaxation time.43, 44 Furthermore, nanoscale size enhances phonon-surface scattering,
which could significantly suppress thermal conductivity and hence boost the figure of
merit ZT.44-58
Despite the fact that nanostructured SnTe shows great promise in various aspects, the
surface of SnTe is prone to contamination and oxidation
59-61
which creates difficulty
for the measurement of topological surface states and may also cause degradation in
material performance. Direct evidence of topological surface states in SnTe-based bulk
crystals has been obtained by surface sensitive ARPES measurements on in situ cleaved
surfaces in ultrahigh vacuum.2 Surface examinations on nanostructured SnTe
compounds using nano-ARPES, however, are more difficult due to the lack of effective
cleaning methods for ex situ grown nanostructures and the unavoidable exposure to air
during sample transfer. The critical requirements for accurate surface probing hinder
our understanding of the topological surface states of SnTe compounds at the nanoscale.
Thus, a thorough investigation on the surface oxidation of nanostructured SnTe is
essential, and a search for more effective surface cleaning methods is imperative.
In this work, we carried out surface oxidation, cleaning and thermoelectric studies of
In-doped SnTe nanowires grown by chemical vapor deposition. In contrast to the
catalyst-free growth of nanoplates42, 62 and thin films,63 Au nanoparticles were employed
to catalyze the growth of nanowires. The surface of In-doped SnTe was characterized
by transmission electron microscopy (TEM) and X-ray photoelectron spectroscopy
(XPS) studies, which revealed the formation of In2O3, SnO2, Te and TeO2 due to rapid
oxidation in exposure to atmosphere. The surface oxides can be readily removed by
gentle argon ion sputtering, yielding a potential approach to cleaning the surface of ex
situ grown nanostructures for surface sensitive measurements such as ARPES and
scanning tunneling microscopy (STM). Characterizations of electrical conductivity σ,
thermopower S, and thermal conductivity κ were performed on the same In-doped SnTe
nanowire which shows suppressed σ and κ but enhanced S rendering an improved
thermoelectric figure of merit ZT in comparison to the undoped SnTe.
Experimental Details
1. Synthesis
In-doped SnTe nanowires were synthesized inside a three-zone tube furnace (Figure 1a)
via a chemical vapor deposition process that was used for the growth of undoped SnTe
nanowires.37-39 Tin telluride powder (Alfa Aesar, purity 99.999%) was placed in a
combustion boat located in the center of the tube furnace, indium powder (Alfa Aesar,
purity 99.99%) and tellurium powder (Alfa Aesar, purity 99.999%) were placed in the
upstream position 8 inches away from the center. Three ~3-inch long silicon substrates
coated with gold colloids (Ted Pella, diameter 80 nm) were placed adjacent to each other
4
in the downstream position from 4 inches to 13 inches away from the center of the tube
furnace. Ultra-high purity argon, as a carrier gas, was introduced and kept at a flow rate
of 10 sccm with a pressure maintained at 30 Torr. During the growth of nanowires,
temperatures of three zones in the tube furnace were kept at 480/600/475 ˚C (set points)
for 5 hours (see the main text for the actual growth temperatures measured using an
external thermocouple). After the growth was complete, the heater was turned off and
the system was cooled to room temperature over about 5 hours.
2. Characterization
The morphology of nanostructures was characterized by a scanning electron microscope
(SEM, Quanta FEI). The crystal structure and orientation were characterized by TEM
(FEI Tecnai F30). Raman measurements were performed in a confocal Raman
microscope system (Renishaw inVia) using a 532 nm excitation laser under ambient
environment. The chemical composition was determined by Energy-dispersive X-ray
Spectroscopy (EDX) in SEM. XPS measurements were conducted using a PHI Versa
Probe II instrument. X-ray beam with a power of 50 W at 15 kV and a spot size of 200
μm was generated by a monochromatic Al 𝑘𝛼 source. The calibration of instrument
work function was carried out using a standard procedure described previously. 64 For
all samples, neutralization was provided by using the PHI dual charge compensation
system. XPS spectra were taken at the energy step of 0.1 eV using SmartSoft–XPS
v2.3.1 at pass energies of 46.95 eV for Te 3d, 23.5 eV for C 1s, 11.75 eV for Sn 3d, and
93.9 eV for In 3d transitions. The spectra were calibrated using either highly oriented
pyrolytic graphite or carbon 1s peak and were further analyzed in PHI MultiPack v9.0
software. Peaks were fitted using GL line shapes with 80% Gaussians and 20%
Lorentzians, along with a Shirley background.
3. Device fabrication and electrical measurements
The thermoelectric devices were fabricated and measured using the approaches reported
earlier.65-68 The thermoelectric platform with four electrodes and a resistive heater was
fabricated on top of the SiNx/SiO2/Si substrate by e-beam lithography (EBL) followed
by metal deposition of ~ 10 nm Ti / 90 nm Au. A trench was etched in the middle region
of the platform to limit heat transfer on the surface of the substrate. An In-doped SnTe
nanowire was picked up by a nanomanipulator, and then placed on the platform. EBL
and metal deposition (~ 10 nm Ti /290 nm Au) was performed again to connect the
nanowire with the electrodes on the platform. Right before metal deposition, the sample
was cleaned by oxygen plasma and then soaked in 1% HCl for ~80 seconds to remove
remaining e-beam resist and oxides on the surface of the nanowire. The electrical
conductivity was determined by a standard four-probe measurement of resistance. The
thermopower was measured by applying a current through the heater, creating a
temperature gradient along the nanowire. The thermoelectric voltage Vth between B and
C were measured. Electrodes B and C also serve as thermometers by measuring their
resistances, and the temperature difference ΔT between B and C was obtained.
Thermopower of the device was then calculated by 𝑆 = 𝑉𝑡ℎ /∆𝑇, and the thermopower
of the nanowire was determined by subtracting contribution from the Au electrodes. The
thermal conductivity was measured using a self-heating method: by applying a large
current through electrodes A and D, a voltage between B and C was measured and the
resistance was calculated.65-69 The average temperature increase of the nanowire ∆TM
was then determined from the resistance change (a linear temperature dependence is
assumed in a 10 K interval). The thermal conductivity can be calculated based on 𝜅 =
𝐼2 𝑅𝑙/(12Δ𝑇𝑀 𝐴), where I is the current, R is the resistance, l is the channel length, and
A is the cross section of the nanowire.65-69
6
Results and Discussion
Figure 1a shows a schematic of the chemical vapor deposition system for nanowire
synthesis. Using Au as catalyst, indium, tellurium and tin telluride as precursors, we
achieved the growth of In-doped SnTe nanowires with a variety of morphologies. A
high density of straight and smooth nanowires with a typical width of 100 – 200 nm and
length of about 5 – 8 μm were observed on the first substrate (closest to the precursors)
in a temperature range of 605 ˚C ~ 525 ˚C (Figure 1b). On the second substrate (further
away from the precursors) where the temperature is 525 ˚C ~ 487 ˚C, we observed large,
smooth nanobeams with a typical width of about 1 μm and length of 20 – 50 μm (Figure
1c). When the temperature is further reduced on the third substrate (farthest from the
precursors) to 487 ˚C ~ 359 ˚C, ‘twisted’ nanowires are formed (Figure 1d). The typical
width of the ‘twisted’ nanowires varies from about 100 nm to 300 nm and the length is
1 – 3 μm. The insets of Figures 1b-d are the corresponding high-magnification SEM
images taken near the tip of the nanowires. Au-Sn-Te alloy particle is observed on the
tip of these nanowires/nanobeams, indicating a vapor-liquid-solid growth mechanism
(Figure S1, Supplementary Information). The nanowire diameter is usually larger than
the particle, suggesting direct vapor-solid deposition on the side wall of the nanowire.
Along with the nanowires are a high density of crystals whose size ranges from
nanometers to micrometers. Since the dimension and geometry of the nanowires grown
on the first substrate are the most suitable for thermoelectric measurement, our
discussions below will focus on them unless otherwise noted.
The high crystalline quality of nanowires was demonstrated by TEM. Figure 2a
shows a TEM image taken on the body of a typical nanowire from the first substrate,
the nanowire surface is smooth and the width of the nanowire is about 104 nm. The
selected area electron diffraction (SAED) pattern is shown in Figure 2b, which suggests
the single crystallinity of the nanowire and the growth direction along [100]. A high
resolution TEM (HRTEM) image is shown in Figure 2c, the d-spacing between crystal
planes perpendicular to the growth direction is measured to be about 0.32 nm,
corresponding to the distance between two adjacent (200) planes. Fast Fourier transform
(FFT) was carried out near the edge and the core of the nanowire body respectively. The
edge of the nanowire body appears to be amorphous with a circular FFT pattern absent
of distinguishable diffraction spots as shown in Figure 2d. Whereas FFT taken on the
core has clear diffraction spots indicating the crystalline nature (Figure 2e).
The
distinction in crystallinity between the edge and the core of the nanowire is likely caused
by surface oxidation of the compound. A Raman spectrum taken on a typical nanowire
shows two prominent peaks at ~ 124 cm-1 and ~142 cm-1 (Figure 2f). While all phonon
modes are Raman inactive in a rock-salt crystal structure, these peaks were also
observed in both undoped and Mn-doped SnTe bulk crystals, which was attributed to
the presence of defects (e.g. Sn vacancies) or the breaking of symmetry on the surface.7072
To characterize the bulk and surface compositions of the nanowires, we carried out
EDX and XPS measurements. The EDX examines the bulk of the sample, while XPS
probes the surface layers (~10 nm deep) with a higher sensitivity. In addition to
composition, XPS also characterizes the oxidation state of each element. As shown in
Figure S2 (Supplementary Information), Sn and Te peaks are observed in the EDX
spectra while In is not, indicating that the dopant concentration is below the detection
limit. The Sn:Te ratio varies slightly from one nanowire to another but is close to ~1
within the accuracy of the measurement. XPS spectra taken on a pristine sample (i.e.
before sputtering as discussed below) reveal clearly the presence of oxides on the
surface. A strong oxygen 1s peak was observed in Figure 3a where one component with
8
a higher binding energy corresponds to -2 state in SiO2 from the substrate, and the other
components are related to -2 state in oxides from SnTe surface. The Sn 3d spectrum in
Figure 3b shows two peaks at ~ 496.0 eV (3d3/2) and ~ 487.6 eV (3d5/2), indicating a +4
oxidation state in contrast to the nominal +2 in SnTe.59-61 The Te 3d spectrum in Figure
3c has two components with binding energies of 583.9 eV (3d 3/2), 573.5 eV (3d5/2) and
587.8 eV (3d3/2), 577.4 eV (3d5/2), respectively. The former corresponds to a 0 state,73
and the latter is +4 as in TeO2.59-61 In contrast to the EDX characterization, two
prominent In peaks are observed in Figure 3d and the binding energies are 453.7 eV
(3d3/2) and 446.0 eV (3d5/2) which correspond to a +3 oxidation state.74 These results
suggest that the sample surface is oxidized and forms In 2O3, SnO2, Te and TeO2. The
atomic ratio determined based on the XPS spectra is In: Sn: Te=3.3:77.4:19.3. XPS
spectra taken at other spots on the first substrate show similar results (Figure S3 and
Table S1, Supplementary Information). The large deviation of Sn:Te ratio from the
nominal value in SnTe was previously observed in undoped SnTe bulk crystals 60 and
thin films59 as well, and is attributed to the different oxidation rates of the two elements.
XPS spectra were also taken on the second and the third substrates (Figure S4,
Supplementary Information). On the second substrate, the Sn 3d and Te 3d core level
spectra are qualitatively similar to those on the first substrate. However, no In peak was
observed, indicating that the indium concentration on the second substrate is below the
detection limit. The atomic ratio determined from the XPS spectra taken at one of the
three spots (Table S2, Supplementary Information) is Sn:Te=82.3:17.7, again
suggesting a surface with excessive oxides of Sn. On the third substrate, strong In 3d
peaks were observed, and quantitative analysis of a representative set of data shows an
atomic ratio of In: Sn: Te=9.1:41.6:49.3 (Figure S5 and Table S3, Supplementary
Information). The enrichment of In and Te indicates the possible formation of indium
telluride near the crystal surface caused by the lower growth temperature of the third
substrate.
The presence of surface oxides is a critical challenge to study the topological states
via surface sensitive techniques such as (nano)-ARPES and STM. Therefore, it is
important to develop an effective approach to removing the surface oxides. We
demonstrate here that argon sputtering can be used to achieve this purpose. Figure 3
compares the XPS spectra before and after sputtering. The sputtering process was
performed for 1, 2, and 3 hours. After 1 hour of sputtering, the oxygen 1s peak is
significantly reduced (Figure 3a). In the meantime, the Sn 4+ peaks disappeared,
accompanied by the appearance of Sn2+ peaks with lower binding energies of 494.1 eV
and 485.6 eV (Figure 3b). The Te4+ and Te peaks also disappeared, while the Te2- peaks
became visible with binding energies of 583.2 eV and 572.7 eV (Figure 3c). This result
suggests that surface SnO2, Te and TeO2 were removed by sputtering. Upon the removal
of surface In2O3, the In 3d peaks were shifted to lower binding energies and the peak
intensities were reduced. The atomic ratio of In: Sn: Te after 1 hour of oxide removal
becomes: 0.6:55.1:44.3, corresponding to an indium doping concertation of ~ 1%. The
XPS spectra do not change notably under further sputtering, and the atomic ratio
determined from the spectra does not show a strong variation as a function of sputtering
time (Table 1). We therefore conclude that the sample composition is quite uniform in
the non-oxidized region.
We further carried out re-oxidation experiments to demonstrate that surface oxidation
occurs rapidly in atmospheric environment. Figures 4a-c compare the XPS spectra taken
right after the sample was cleaned by sputtering for 30 minutes and after it was exposed
to air for 1, 2 and 4 mins, respectively. Right after sputtering, the Sn and Te are mainly
in +2 and -2 states, respectively. The slight asymmetry of the Te peaks is caused by a
10
small amount (~15%) of residual Te due to incomplete cleaning. After only 1 minute of
exposure, Sn4+ peaks appear and dominate over the Sn2+; the Te4+ peaks are also
observed in spite of a weak intensity. The oxygen 1s peak was significantly enhanced
accordingly. The change in spectra is relatively small under further exposure for 2 and
4 minutes. To obtain a quantitative picture, we fitted the XPS spectra of Sn 3d 5/2 using
two components (Sn4+ and Sn2+) and Te 3d5/2 using three components (Te4+, Te and Te2) and plot the Sn4+/Sntotal and Te4+/ Tetotal ratios as a function of exposure time in Figure
4d. It is clear that the oxidation occurs rapidly within 1 minute of exposure and then the
oxidation rate decreases significantly. Also in Table 1, the atomic ratio determined from
the spectra taken after 1 minute of exposure to atmosphere shows a sudden change
compared to that obtained immediately after sputtering, but longer exposures of 2 and 4
minutes do not induce further significant changes in atomic ratio. This suggests that
substantial oxidation occurs within 1 minute of exposure.
Lastly, we demonstrate the enhanced thermoelectric figure of merit of In-doped SnTe
nanowires by measuring the electrical conductivity, thermopower and thermal
conductivity of the same nanowire. The detailed information of device fabrication and
measurment is provided in the Experimental Section. The electrical conductivity σ
determined from a standard four-terminal measurement is ~1.49 × 105 S m-1 at 300 K,
which is close to the conductivity of In-doped SnTe polycrystalline samples (~2.0 – 4.3
× 105 S m-1) 8 and nanoplates (~1.4 × 105 S m-1),62 but is clearly lower than that of the
undoped SnTe nanowires (6 – 7 × 105 S m-1).66 The σ increases when the sample is
cooled down to around 100 K and then starts to decrease, rendering a metalsemiconductor transition which is in contrast to the pure metallic behavior observed in
the undoped SnTe.66 The metallic behavior in SnTe is attributed to degenerate doping
by native Sn vacancies which act as acceptors (or p-type dopants). The density of Sn
vacancies is estimated to be on the order of 1020 cm-3 (or 1%),19, 66 comparable to the
indium dopant concentration in the core of the wire. While indium is considered as a ptype dopant as well, it provides less holes than a Sn vacancy. 8 So when In atoms fill in
the Sn vacancies, the hole density decreases. Another influence of indium doping is the
introduction of disorder, which reduces hole mobility by enhancing carrier-impurity
scattering. Both the reduction of hole density and mobility could be responsible for the
suppression of electrical conductivity upon doping.
Despite its suppressed electrical conductivity, the In-doped SnTe nanowire shows
enhanced thermopower S in comparison to the undoped SnTe. The temperature
dependence of S is shown in Figure 5b. At room temperature, the thermopower is 59 µV
K-1, which is comparable with the values measured on the polycrystalline samples (~50
– 75 µV K-1)8 and is enhanced relative to the undoped SnTe nanowires (19 – 41 µV K1 66
).
The positive sign of S confirms the p-type nature of the majority charge carriers.
The thermopower decreases monotonically with the decrease of temperature and shows
no bipolar effect. The total thermal conductivity κ of the nanowire was measured using
a self-heating method.65-69 As shown in Figure 5c, the κ at 300 K is 2.24 W m-1 K-1,
slightly lower than that of both the In-doped SnTe pollycrystal (~3.8 – 4.4 W m-1 K-1)8
and the undoped SnTe nanowires (~4.2 W m-1 K-1).66 The total thermal conductivity can
be decomposed into two parts: electronic contribution and lattice contribution. The
electronic thermal conductivity κe is determined based on the Wiedemann-Franz law,
i.e. 𝜅𝑒 = 𝐿𝜎𝑇, where L is the Lorenz number. We calculated the Lorentz number using
the same model for bulk SnTe in which contributions from both the light hole band and
heavy hole band are included.8 The calculated Lorenz number (e.g. 1.65 × 10-8 V2 K-2 at
300 K) falls in the range of ~1.49 × 10-8 V2 K-2 for nondegenerate semiconductors and
~2.44 × 10-8 V2 K-2 for metals.75 The lattice thermal conductivity κL is then determined
12
by subtracting the total thermal conductivity by the electronic thermal conductivity.
Both κe and κL are plotted in Figure 5c, κe increases as the temperature is raised, while
κL shows a non-monotonic behavior with a maximum value at 60K. The non-monotonic
relationship between κL and temperature may be due to the increase of phonon density
upon warming at lower temperatures and the enhancement of umklapp phonon-phonon
scattering at higher temperatures.66 The ZT was finally calculated and plotted as a
function of temperature in Figure 5d. At 300 K, the ZT of our nanowire is 0.07, which
is comparable to that of In-doped SnTe polycrystalline samples (~ 0.08 – 0.09)8 but
higher than that of undoped SnTe nanowires (~0.0379).66 The enhancement of ZT in the
indium-doped nanowire should be attributed to both the improved thermopower and the
suppressed thermal conductivity.
Conclusions
We presented the first growth of In-doped SnTe nanowire via chemical vapor
deposition. The nanowires were grown under the assistance of Au catalysts, with various
morphologies depending on the substrate position and growth temperature. The surface
of the nanowires is amorphous contrary to the single crystalline core as revealed by
transmission electron microscopy. The surface amorphous layer was further studied by
X-ray photoelectron spectroscopy which suggested its composition to be In 2O3, SnO2,
Te and TeO2. Argon ion sputtering was found to effectively remove the amorphous
layer. We also demonstrated that exposure of the cleaned nanowires to air within one
minute led to rapid oxidation of the surface. Compared to undoped SnTe, an enhanced
thermoelectric figure of merit ZT was obtained from our nanowires doped by ~1% of
indium. The enhancement is attributed to the suppression of thermal conductivities and
the improvement of thermopower.
Acknowledgements
We are grateful for the experimental assistance provided by Dr. Julio Martinez, John
Nogan, Anthony R. James, Douglas V. Pete, Denise B. Webb, and Renjie Chen. We
acknowledge support, in part, from LDRD program at Los Alamos National Laboratory,
start-up funding at Indiana University and NSF grant via DMR-1506460. We thank the
Nanoscale Characterization Facility at Indiana University for instrumentation access.
The XPS instrument was funded by NSF Award DMR MRI-1126394. This work was
performed, in part, at the Center for Integrated Nanotechnologies, an Office of Science
User Facility operated for the U.S. Department of Energy (DOE) Office of Science by
Los Alamos National Laboratory (Contract DE-AC52- 06NA25396) and Sandia
National Laboratories (Contract DE-AC04-94AL85000).
References
1
T. H. Hsieh, H. Lin, J. Liu, W. Duan, A. Bansil and L. Fu, Nat. Commun., 2012, 3, 982.
2
Y. Tanaka, Z. Ren, T. Sato, K. Nakayama, S. Souma, T. Takahashi, K. Segawa and Y.
Ando, Nat. Phys., 2012, 8, 800-803.
3
P. Dziawa, B. J. Kowalski, K. Dybko, R. Buczko, A. Szczerbakow, M. Szot, E.
Lusakowska, T. Balasubramanian, B. M. Wojek, M. H. Berntsen, O. Tjernberg and T.
Story, Nat. Mater., 2012, 11, 1023-1027.
4
S. Y. Xu, C. Liu, N. Alidoust, M. Neupane, D. Qian, I. Belopolski, J. D. Denlinger, Y. J.
Wang, H. Lin, L. A. Wray, G. Landolt, B. Slomski, J. H. Dil, A. Marcinkova, E. Morosan,
14
Q. Gibson, R. Sankar, F. C. Chou, R. J. Cava, A. Bansil and M. Z. Hasan, Nat. Commun.,
2012, 3, 1192.
5
Y. Okada, M. Serbyn, H. Lin, D. Walkup, W. Zhou, C. Dhital, M. Neupane, S. Xu, Y. J.
Wang, R. Sankar, F. Chou, A. Bansil, M. Z. Hasan, S. D. Wilson, L. Fu and V. Madhavan,
Science, 2013, 341, 1496-1499.
6
I. Zeljkovic, Y. Okada, C.-Y. Huang, R. Sankar, D. Walkup, W. Zhou, M. Serbyn, F. Chou,
W.-F. Tsai, H. Lin, A. Bansil, L. Fu, M. Z. Hasan and V. Madhavan, Nat. Phys., 2014, 10,
572-577.
7
P. B. Littlewood, B. Mihaila, R. K. Schulze, D. J. Safarik, J. E. Gubernatis, A. Bostwick,
E. Rotenberg, C. P. Opeil, T. Durakiewicz, J. L. Smith and J. C. Lashley, Physical Review
Letters, 2010, 105, 086404.
8
Q. Zhang, B. Liao, Y. Lan, K. Lukas, W. Liu, K. Esfarjani, C. Opeil, D. Broido, G. Chen
and Z. Ren, Proc. Natl. Acad. Sci. U.S.A., 2013, 110, 13261-13266.
9
G. Tan, L. D. Zhao, F. Shi, J. W. Doak, S. H. Lo, H. Sun, C. Wolverton, V. P. Dravid, C.
Uher and M. G. Kanatzidis, J. Am. Chem. Soc., 2014, 136, 7006-7017.
10 M. Zhou, Z. M. Gibbs, H. Wang, Y. Han, C. Xin, L. Li and G. J. Snyder, Phys. Chem.
Chem. Phys., 2014, 16, 20741-20748.
11 L. Fu, Phys. Rev. Lett., 2011, 106, 106802.
12 J. Liu, W. Duan and L. Fu, Phys. Rev. B, 2013, 88, 241303(R).
13 G. A. Fiete, Nat. Mater., 2012, 11, 1003-1004.
14 I. Zeljkovic, Y. Okada, M. Serbyn, R. Sankar, D. Walkup, W. Zhou, J. Liu, G. Chang, Y.
J. Wang, M. Z. Hasan, F. Chou, H. Lin, A. Bansil, L. Fu and V. Madhavan, Nat. Mater.,
2015, 14, 318-324.
15 I. Zeljkovic, D. Walkup, B. A. Assaf, K. L. Scipioni, R. Sankar, F. Chou and V. Madhavan,
Nat. Nanotechnol., 2015, 10, 849-853.
16 A. S. Erickson, J. H. Chu, M. F. Toney, T. H. Geballe and I. R. Fisher, Phys. Rev. B, 2009,
79, 024520.
17 G. Balakrishnan, L. Bawden, S. Cavendish and M. R. Lees, Phys. Rev. B, 2013, 87,
140507(R).
18 R. D. Zhong, J. A. Schneeloch, X. Y. Shi, Z. J. Xu, C. Zhang, J. M. Tranquada, Q. Li and
G. D. Gu, Phys. Rev. B, 2013, 88, 020505(R).
19 M. Novak, S. Sasaki, M. Kriener, K. Segawa and Y. Ando, Phys. Rev. B, 2013, 88,
140502(R).
20 V. K. Maurya, Shruti, P. Srivastava and S. Patnaik, Europhys. Lett., 2014, 108, 37010.
21 T. Sato, Y. Tanaka, K. Nakayama, S. Souma, T. Takahashi, S. Sasaki, Z. Ren, A. A. Taskin,
K. Segawa and Y. Ando, Phys. Rev. Lett., 2013, 110, 206804.
22 A. Y. Kitaev, Phys. Usp., 2001, 44, 131-136.
23 A. Banik and K. Biswas, J. Mater. Chem. A, 2014, 2, 9620-9625.
24 G. Tan, F. Shi, J. W. Doak, H. Sun, L.-D. Zhao, P. Wang, C. Uher, C. Wolverton, V. P.
Dravid and M. G. Kanatzidis, Energy Environ. Sci., 2015, 8, 267-277.
25 A. Banik, U. S. Shenoy, S. Anand, U. V. Waghmare and K. Biswas, Chem. Mater., 2015,
27, 581-587.
26 G. Tan, F. Shi, S. Hao, H. Chi, L.-D. Zhao, C. Uher, C. Wolverton, V. P. Dravid and M. G.
Kanatzidis, J. Am. Chem. Soc., 2015, 137, 5100-5112.
27 G. Tan, F. Shi, S. Hao, H. Chi, T. P. Bailey, L.-D. Zhao, C. Uher, C. Wolverton, V. P.
Dravid and M. G. Kanatzidis, J. Am. Chem. Soc., 2015, 137, 11507-11516.
28 H. Wu, C. Chang, D. Feng, Y. Xiao, X. Zhang, Y. Pei, L. Zheng, D. Wu, S. Gong, Y. Chen,
J. He, M. G. Kanatzidis and L.-D. Zhao, Energy Environ. Sci., 2015, 8, 3298-3312.
29 L.-D. Zhao, X. Zhang, H. Wu, G. Tan, Y. Pei, Y. Xiao, C. Chang, D. Wu, H. Chi, L. Zheng,
S. Gong, C. Uher, J. He and M. G. Kanatzidis, J. Am. Chem. Soc., 2016, 138, 2366-2373.
16
30 D. H. Damon, J. Appl. Phys., 1966, 37, 3181-3190.
31 M. Zhou, Z. M. Gibbs, H. Wang, Y. Han, L. Li and G. J. Snyder, Appl. Phys. Lett., 2016,
109, 042102.
32 M.-K. Han, J. Androulakis, S.-J. Kim and M. G. Kanatzidis, Adv. Energy Mater., 2012, 2,
157-161.
33 Y. Chen, M. D. Nielsen, Y.-B. Gao, T.-J. Zhu, X. Zhao and J. P. Heremans, Adv. Energy
Mater., 2012, 2, 58-62.
34 X. Shi, J. R. Salvador, J. Yang and H. Wang, Sci. Adv. Mater., 2011, 3, 667-671.
35 V. P. Vedeneev, S. P. Krivoruchko and E. P. Sabo, Semiconductors, 1998, 32, 241-244.
36 M. M. Asadov, M. A. Alidzhanov, F. M. Mamedov and G. I. Kelbaliev, Inorg. Mater.,
1998, 34, 442-444.
37 Z. Li, S. Shao, N. Li, K. McCall, J. Wang and S. X. Zhang, Nano Lett., 2013, 13, 54435448.
38 M. Safdar, Q. Wang, M. Mirza, Z. Wang, K. Xu and J. He, Nano Lett., 2013, 13, 53445349.
39 M. Saghir, M. R. Lees, S. J. York and G. Balakrishnan, Cryst. Growth Des., 2014, 14, 20092013.
40 J. Shen, Y. Jung, A. S. Disa, F. J. Walker, C. H. Ahn and J. J. Cha, Nano Lett., 2014, 14,
4183-4188.
41 Q. Wang, M. Safdar, Z. Wang, X. Zhan, K. Xu, F. Wang and J. He, Small, 2015, 11, 20192025.
42 J. Shen, Y. Xie and J. J. Cha, Nano Lett., 2015, 15, 3827-3832.
43 Y. Xu, Z. Gan and S.-C. Zhang, Phys. Rev. Lett., 2014, 112, 226801.
44 M. Hong, Z.-G. Chen, L. Yang and J. Zou, Nanoscale, 2016, 8, 8681-8686.
45 D. Li, Y. Wu, P. Kim, L. Shi, P. Yang and A. Majumdar, Appl. Phys. Lett., 2003, 83, 29342936.
46 A. I. Hochbaum, R. Chen, R. D. Delgado, W. Liang, E. C. Garnett, M. Najarian, A.
Majumdar and P. Yang, Nature, 2008, 451, 163-167.
47 A. I. Boukai, Y. Bunimovich, J. Tahir-Kheli, J.-K. Yu, W. A. Goddard III and J. R. Heath,
Nature, 2008, 451, 168-171.
48 A. L. Moore, M. T. Pettes, F. Zhou and L. Shi, J. Appl. Phys., 2009, 106, 034310.
49 J. W. Roh, S. Y. Jang, J. Kang, S. Lee, J.-S. Noh, W. Kim, J. Park and W. Lee, Appl. Phys.
Lett., 2010, 96, 103101.
50 P. N. Martin, Z. Aksamija, E. Pop and U. Ravaioli, Nano Lett., 2010, 10, 1120-1124.
51 J. Kang, J. W. Roh, W. Shim, J. Ham, J.-S. Noh and W. Lee, Adv. Mater., 2011, 23, 34143419.
52 J. R. Szczech, J. M. Higgins and S. Jin, J. Mater. Chem., 2011, 21, 4037-4055.
53 E. K. Lee, L. Yin, Y. Lee, J. W. Lee, S. J. Lee, J. Lee, S. N. Cha, D. Whang, G. S. Hwang,
K. Hippalgaonkar, A. Majumdar, C. Yu, B. L. Choi, J. M. Kim and K. Kim, Nano Lett.,
2012, 12, 2918-2923.
54 C. T. Bui, R. Xie, M. Zheng, Q. Zhang, C. H. Sow, B. Li and J. T. L. Thong, Small, 2012,
8, 738-745.
55 G. Li, D. Liang, R. L. J. Qiu and X. P. A. Gao, Appl. Phys. Lett., 2013, 102, 043104.
56 J. Kim, S. Lee, Y. M. Brovman, P. Kim and W. Lee, Nanoscale, 2015, 7, 5053-5059.
57 H. Tang, X. Wang, Y. Xiong, Y. Zhao, Y. Zhang, Y. Zhang, J. Yang and D. Xu, Nanoscale,
2015, 7, 6683-6690.
58 Z. Li, Y. Chen, J.-F. Li, H. Chen, L. Wang, S. Zheng and G. Lu, Nano Energy, 2016, 28,
78-86.
59 M. Bettini and H. J. Richter, Surf. Sci., 1979, 80, 334-343.
18
60 V. S. Neudachina, T. B. Shatalova, V. I. Shtanov, L. V. Yashina, T. S. Zyubina, M. E.
Tamm and S. P. Kobeleva, Surf. Sci., 2005, 584, 77-82.
61 M. Saghir, M. Walker, C. F. McConville and G. Balakrishnan, Appl. Phys. Lett., 2016, 108,
061602.
62 S. Sasaki and Y. Ando, Cryst. Growth Des., 2015, 15, 2748-2752.
63 W. Si, C. Zhang, L. Wu, T. Ozaki, G. Gu and Q. Li, Appl. Phys. Lett., 2015, 107, 092601.
64 R. Easterday, O. Sanchez-Felix, Y. Losovyj, M. Pink, B. D. Stein, D. G. Morgan, M.
Rakitin, V. Y. Doluda, M. G. Sulman, W. E. Mahmoud, A. A. Al-Ghamdi and L. M.
Bronstein, Catal. Sci. Technol., 2015, 5, 1902-1910.
65 J. A. Martinez, P. P. Provencio, S. T. Picraux, J. P. Sullivan and B. S. Swartzentruber, J.
Appl. Phys., 2011, 110, 074317.
66 E. Z. Xu, Z. Li, J. A. Martinez, N. Sinitsyn, H. Htoon, N. Li, B. Swartzentruber, J. A.
Hollingsworth, J. Wang and S. X. Zhang, Nanoscale, 2015, 7, 2869-2876.
67 E. Xu, Z. Li, J. A. Acosta, N. Li, B. Swartzentruber, S. Zheng, N. Sinitsyn, H. Htoon, J.
Wang and S. Zhang, Nano Res., 2016, 9, 820-830.
68 J. A. Martinez, J.-H. Cho, X. Liu, T. S. Luk, J. Huang, S. T. Picraux, J. P. Sullivan and B.
S. Swartzentruber, Appl. Phys. Lett., 2013, 102, 103101.
69 F. Völklein, H. Reith, T. W. Cornelius, M. Rauber and R. Neumann, Nanotechnology,
2009, 20, 325706.
70 S. Sugai, K. Murase, S. Katayama, S. Takaoka, S. Nishi and H. Kawamura, Solid State
Commun., 1977, 24, 407-409.
71 C. An, K. Tang, B. Hai, G. Shen, C. Wang and Y. Qian, Inorg. Chem. Commun., 2003, 6,
181-184.
72 S. Acharya, J. Pandey and A. Soni, Appl. Phys. Lett., 2016, 109, 133904.
73 A. J. Ricco, H. S. White and M. S. Wrighton, J. Vac. Sci. Technol. A, 1984, 2, 910-915.
74 W. K. Liu, W. T. Yuen and R. A. Stradling, J. Vac. Sci. Technol. B, 1995, 13, 1539-1545.
75 K. T. Wojciechowski and M. Schmidt, Phys. Rev. B, 2009, 79, 184202.
20
Figure 1 (a) Schematic picture of the chemical vapor deposition system used for the growth of
nanostructures. The small quartz tubes that hold individual crucibles and substrates are not
shown in the picture. SEM images taken on the (b) first substrate (closest to the precursors),
(c) second substrate (further away from the precursors) and (d) third substrate (farthest from
the precursors). Insets are magnified SEM images on the tip of a typical nanowire from (b) first
substrate, (c) second substrate and (d) third substrate respectively.
Figure 2 (a) Low resolution TEM image of a typical In-doped SnTe nanowire after prolonged
exposure to air. (b) SAED pattern of the nanowire. (c) HRTEM image of the nanowire body.
Red frames indicate the location where FFT was conducted. (d) FFT taken near the edge of the
nanowire. (e) FFT taken near the core of the nanowire. (f) Raman spectrum from a typical
nanowire.
22
Figure 3 XPS spectra: (a) O 1s, (b) Sn 3d, (c) Te 3d and (d) In 3d core level spectra taken
immediately before sputtering and after 1, 2 and 3 hours of sputtering respectively. All spectra
were calibrated using highly oriented pyrolytic graphite due to the absence of carbon 1s peak
after sputtering. For the spectra taken before sputtering, calibration using carbon 1s peak yields
slightly lower binding energies. The dashed lines indicate roughly the peaks corresponding to
different oxidation states. The spectra are vertically shifted for clarity.
Figure 4 XPS spectra: (a) O 1s, (b) Sn 3d and (c) Te 3d core level spectra taken after sputtering,
after 1, 2 and 4 minutes of exposure in atmosphere respectively. The spectra were calibrated
using carbon 1s peak and were vertically shifted for clarity. The dashed lines indicate roughly
the peaks corresponding to different oxidation states. (d) Atomic ratio of Sn4+/Sntotal and Te4+/
Tetotal as a function of exposure time.
24
Figure 5 (a) SEM image of an In-doped SnTe device for thermoelectric measurement. (b)
Electrical conductivity and thermopower as a function of temperature. (c) Total thermal
conductivity κ, electronic thermal conductivity κe and lattice thermal conductivity κL versus
temperature. (d) ZT plotted against temperature.
Sputtering
treatment
In / Sn / Te
Re-oxidation
In / Sn / Te
Before sputtering
3.3/77.4/19.3
After sputtering
0.7/58.7/40.6
Sputtered for 1 h
0.6/55.1/44.3
Sputtered for 2 h
0.7/56.0/43.3
Sputtered for 3 h
0.4/56.7/42.8
Exposed for 1
min
Exposed for 2
min
Exposed for 4
min
0.8/66.7/32.5
0.6/68.8/30.7
0.6/67.9/31.6
Table 1 Sample compositions after sputtering treatment and re-oxidation in atmosphere
26